class TestGffWriter:
RECORD1 = Gff3Record("chr1",
10,
11,
"insertion",
attributes=[("cat", "1"), ("dog", "2")])
RECORD2 = Gff3Record("chr1",
200,
201,
"substitution",
attributes=[("mouse", "1"), ("moose", "2")])
def setup_method(self):
self.outfile = StringIO()
self.gffWriter = GffWriter(self.outfile)
def test_writeHeader(self):
self.gffWriter.writeHeader("##foo bar")
assert "##gff-version 3\n##foo bar\n" == self.outfile.getvalue()
def test_writeRecord(self):
self.gffWriter.writeRecord(self.RECORD1)
self.gffWriter.writeRecord(self.RECORD2)
expected = (
"##gff-version 3\n" +
"chr1\t.\tinsertion\t10\t11\t.\t.\t.\tcat=1;dog=2\n" +
"chr1\t.\tsubstitution\t200\t201\t.\t.\t.\tmouse=1;moose=2\n")
assert expected == self.outfile.getvalue()
def m5CgffConsumer( self, filename ):
f = self.openWriteHandle( filename )
gff = GffWriter( f )
# write headers describing the program that generated the data
gff.writeHeader('##source ipdSummary v2.0')
gff.writeHeader('##source-commandline %s' % self.options.cmdLine)
# Write the reference renaming info into the gff headers ala evicons
# for entry in self.refInfo:
# gff.writeHeader("##sequence-region %s 1 %d"
# % (entry.Name, entry.Length))
try:
while True:
# Pull in a single record?
siteObsList = (yield)
for siteObs in siteObsList:
if 'Ca5C' in siteObs and siteObs['strand'] == 0:
gff.writeRecord( self.makeM5CgffRecord( siteObs ) )
except GeneratorExit:
f.close()
return
def m5CgffConsumer( self, filename ):
f = self.openWriteHandle( filename )
gff = GffWriter( f )
# write headers describing the program that generated the data
gff.writeHeader('##source ipdSummary.py v2.0')
gff.writeHeader('##source-commandline %s' % self.options.cmdLine)
# Write the reference renaming info into the gff headers ala evicons
# for entry in self.refInfo:
# gff.writeHeader("##sequence-region %s 1 %d"
# % (entry.Name, entry.Length))
try:
while True:
# Pull in a single record?
siteObsList = (yield)
for siteObs in siteObsList:
if siteObs.has_key('Ca5C') and siteObs['strand'] == 0:
gff.writeRecord( self.makeM5CgffRecord( siteObs ) )
except GeneratorExit:
f.close()
return
class TestGffWriter(object):
def setup(self):
self.outfile = StringIO()
self.record1 = Gff3Record("chr1",
10,
11,
"insertion",
attributes=[("cat", "1"), ("dog", "2")])
self.record2 = Gff3Record("chr1",
200,
201,
"substitution",
attributes=[("mouse", "1"), ("moose", "2")])
self.gffWriter = GffWriter(self.outfile)
def test_writeHeader(self):
self.gffWriter.writeHeader("##foo bar")
assert_equal("##gff-version 3\n##foo bar\n", self.outfile.getvalue())
def test_writeRecord(self):
self.gffWriter.writeRecord(self.record1)
self.gffWriter.writeRecord(self.record2)
expected = (
"##gff-version 3\n" +
"chr1\t.\tinsertion\t10\t11\t.\t.\t.\tcat=1;dog=2\n" +
"chr1\t.\tsubstitution\t200\t201\t.\t.\t.\tmouse=1;moose=2\n")
assert_equal(expected, self.outfile.getvalue())
class TestGffWriter:
def setup(self):
self.outfile = StringIO()
self.record1 = Gff3Record("chr1", 10, 11, "insertion",
attributes=[("cat", "1"), ("dog", "2")])
self.record2 = Gff3Record("chr1", 200, 201, "substitution",
attributes=[("mouse", "1"), ("moose", "2")])
self.gffWriter = GffWriter(self.outfile)
def test_writeHeader(self):
self.gffWriter.writeHeader("##foo bar")
assert_equal("##gff-version 3\n##foo bar\n",
self.outfile.getvalue())
def test_writeRecord(self):
self.gffWriter.writeRecord(self.record1)
self.gffWriter.writeRecord(self.record2)
expected = ("##gff-version 3\n" +
"chr1\t.\tinsertion\t10\t11\t.\t.\t.\tcat=1;dog=2\n" +
"chr1\t.\tsubstitution\t200\t201\t.\t.\t.\tmouse=1;moose=2\n")
assert_equal(expected, self.outfile.getvalue())
def gffConsumer(self, filename):
"""
Consume IPD summary rows, filter them and write to GFF
"""
#f = file(filename, 'w', 2<<15)
f = self.openWriteHandle(filename)
gff = GffWriter(f)
# write headers describing the program that generated the data
gff.writeHeader('##source ipdSummary v2.0')
gff.writeHeader('##source-commandline %s' % self.options.cmdLine)
# Write the reference renaming info into the gff headers ala evicons
for entry in self.refInfo:
gff.writeHeader("##sequence-region %s 1 %d" %
(entry.Name, entry.Length))
minScore = -10 * math.log10(self.options.pvalue)
snippetRef = -1
try:
while True:
# Pull a record in from the
siteObsList = (yield)
for siteObs in siteObsList:
# self.snippetFunc is a function that return a reference
# snippet given a template position and a strand
if snippetRef != siteObs['refId']:
self.snippetFunc = self.ipdModel.snippetFunc(
siteObs['refId'], 20, 20)
snippetRef = siteObs['refId']
# Two cases for gff entries:
# 1. 'Identified modification' - will have a 'modification' key
# - use the modification name as the gff event type
# - use 'modificationScore' for the gff score
# 2. Detected - no 'modification' key
# - use 'modified_base' as the event type
# - use the single site 'score' property as the gff score
# - do not put this kind into the gff if it contains the a 'offTargetPeak' tag
if siteObs['coverage'] > self.options.minCoverage:
# Case 1
if 'modification' in siteObs and siteObs[
'modification'] != '.':
gff.writeRecord(self.makeGffRecord(siteObs))
# Case 2
elif siteObs[
'score'] > minScore and 'offTargetPeak' not in siteObs:
gff.writeRecord(self.makeGffRecord(siteObs))
# FIXME: Try not filtering:
# gff.writeRecord(self.makeGffRecord(siteObs))
except GeneratorExit:
f.close()
return
def gffConsumer(self, filename):
"""
Consume IPD summary rows, filter them and write to GFF
"""
#f = file(filename, 'w', 2<<15)
f = self.openWriteHandle(filename)
gff = GffWriter(f)
# write headers describing the program that generated the data
gff.writeHeader('##source ipdSummary.py v2.0')
gff.writeHeader('##source-commandline %s' % self.options.cmdLine)
# Write the reference renaming info into the gff headers ala evicons
for entry in self.refInfo:
gff.writeHeader("##sequence-region %s 1 %d" \
% (entry.Name, entry.Length))
minScore = -10*math.log10(self.options.pvalue)
snippetRef = -1
try:
while True:
# Pull a record in from the
siteObsList = (yield)
for siteObs in siteObsList:
# self.snippetFunc is a function that return a reference snippet given a template position and a strand
if snippetRef != siteObs['refId']:
self.snippetFunc = self.ipdModel.snippetFunc(siteObs['refId'],20,20)
snippetRef = siteObs['refId']
# Two cases for gff entries:
# 1. 'Identified modification' - will have a 'modification' key
# - use the modification name as the gff event type
# - use 'modificationScore' for the gff score
# 2. Detected - no 'modification' key
# - use 'modified_base' as the event type
# - use the single site 'score' property as the gff score
# - do not put this kind into the gff if it contains the a 'offTargetPeak' tag
if siteObs['coverage'] > self.options.minCoverage:
# Case 1
if siteObs.has_key('modification') and siteObs['modification'] != '.':
gff.writeRecord(self.makeGffRecord(siteObs))
# Case 2
elif siteObs['score'] > minScore and not siteObs.has_key('offTargetPeak'):
gff.writeRecord(self.makeGffRecord(siteObs))
# FIXME: Try not filtering:
# gff.writeRecord(self.makeGffRecord(siteObs))
except GeneratorExit:
f.close()
return
class VariantsGffWriter(object):
ONTOLOGY_URL = \
"http://song.cvs.sourceforge.net/*checkout*/song/ontology/sofa.obo?revision=1.12"
def __init__(self, f, optionsDict, referenceEntries):
self._gffWriter = GffWriter(f)
self._gffWriter.writeHeader("##pacbio-variant-version 2.1")
self._gffWriter.writeHeader("##date %s" % time.ctime())
self._gffWriter.writeHeader("##feature-ontology %s" %
self.ONTOLOGY_URL)
self._gffWriter.writeHeader("##source GenomicConsensus %s" %
__VERSION__)
self._gffWriter.writeHeader("##source-commandline %s" %
optionsDict["shellCommand"])
self._gffWriter.writeHeader("##source-alignment-file %s" %
optionsDict["inputFilename"])
self._gffWriter.writeHeader("##source-reference-file %s" %
optionsDict["referenceFilename"])
# Reference groups.
for entry in referenceEntries:
self._gffWriter.writeHeader("##sequence-region %s 1 %d" \
% (entry.name, entry.length))
def writeVariants(self, variants):
for var in variants:
self._gffWriter.writeRecord(toGffRecord(var))
def close(self):
self._gffWriter.close()
def gffConsumer(self, filename):
"""
Consume IPD summary rows, filter them and write to GFF
"""
#f = file(filename, 'w', 2<<15)
f = self.openWriteHandle(filename)
gff = GffWriter(f)
# write headers describing the program that generated the data
gff.writeHeader('##source ipdSummary v2.0')
gff.writeHeader('##source-commandline %s' % self.options.cmdLine)
# Write the reference renaming info into the gff headers ala evicons
for entry in self.refInfo:
gff.writeHeader("##sequence-region %s 1 %d" %
(entry.Name, entry.Length))
minScore = -10 * math.log10(self.options.pvalue)
snippetRef = -1
try:
while True:
# Pull a record in from the
siteObsList = (yield)
for siteObs in siteObsList:
true_modif = False
# self.snippetFunc is a function that return a reference snippet given a template position and a strand
if snippetRef != siteObs['refId']:
self.snippetFunc = self.ipdModel.snippetFunc(
siteObs['refId'], 20, 20)
snippetRef = siteObs['refId']
# Two cases for gff entries:
# 1. 'Identified modification' - will have a 'modification' key
# - use the modification name as the gff event type
# - use 'modificationScore' for the gff score
# 2. Detected - no 'modification' key
# - use 'modified_base' as the event type
# - use the single site 'score' property as the gff score
# - do not put this kind into the gff if it contains the a 'offTargetPeak' tag
gffline = self.makeGffRecord(siteObs)
attributes = str(gffline).split()[8].split(';')
real_modif = str(gffline).split()[2]
score_modif = str(gffline).split()[5]
siteObs['type_modif'] = real_modif
siteObs['score_modif'] = score_modif
for attribute in attributes:
key = attribute.split('=')[0]
value = attribute.split('=')[1]
siteObs[str(key)] = value
if siteObs['coverage'] > self.options.minCoverage:
# Case 1
if siteObs.has_key('modification') and siteObs[
'modification'] != '.':
gff.writeRecord(gffline)
true_modif = True
# Case 2
elif siteObs[
'score'] > minScore and not siteObs.has_key(
'offTargetPeak'):
gff.writeRecord(gffline)
true_modif = True
else:
true_modif = False
if true_modif:
siteObs['modified'] = True
else:
siteObs['modified'] = False
siteObs['type_modif'] = 'None'
# Will be printed in sys.stdout for every position (then the pt_methyl or whatsoever will chose it it wants to keep anything)
current_list = []
# exec("siteObs = "+str(siteObs))
for key in siteObs:
current_list.append('{}:'.format(str(key)))
if key != 'capped_values' and key != 'rawData':
for elt in str(siteObs[str(key)]):
if elt != '\n':
current_list.append("{}".format(str(elt)))
else:
current_list.append('None')
current_list.append(';')
# remove the last ';'
current_list.pop(-1)
# Adding the \n
current_list.append('\n')
output = ''.join(current_list)
sys.stdout.write(output)
except GeneratorExit:
f.close()
return
class VariantsGffWriter(object):
ONTOLOGY_URL = \
"http://song.cvs.sourceforge.net/*checkout*/song/ontology/sofa.obo?revision=1.12"
def __init__(self, f, optionsDict, referenceEntries):
self._gffWriter = GffWriter(f)
self._minConfidence = optionsDict["minConfidence"]
self._minCoverage = optionsDict["minCoverage"]
self._gffWriter.writeHeader("##pacbio-variant-version 2.1")
self._gffWriter.writeHeader("##date %s" % time.ctime())
self._gffWriter.writeHeader("##feature-ontology %s" % self.ONTOLOGY_URL)
self._gffWriter.writeHeader("##source GenomicConsensus %s" % __VERSION__)
self._gffWriter.writeHeader("##source-commandline %s" % optionsDict["shellCommand"])
self._gffWriter.writeHeader("##source-alignment-file %s" % optionsDict["inputFilename"])
self._gffWriter.writeHeader("##source-reference-file %s" % optionsDict["referenceFilename"])
# Reference groups.
for entry in referenceEntries:
self._gffWriter.writeHeader("##sequence-region %s 1 %d" \
% (entry.name, entry.length))
def writeVariants(self, variants):
for var in variants:
if var.coverage >= self._minCoverage and var.confidence >= self._minConfidence:
self._gffWriter.writeRecord(toGffRecord(var))
def close(self):
self._gffWriter.close()